The Lancet July 25, 2026

Standard-dose unfractionated heparin versus low-dose unfractionated heparin and low-molecular-weight heparin in extracorporeal life support (RATE): an open-label, randomised, non-inferiority trial

Olivier van Minnen, Annemieke Oude Lansink-Hartgring, Rombout B E van Amstel, et al.

Bottom Line

The RATE trial demonstrated that in adults undergoing ECMO, low-dose unfractionated heparin and therapeutic low-molecular-weight heparin were non-inferior to standard-dose unfractionated heparin for a composite outcome of severe bleeding, severe thromboembolism, or 6-month mortality.

Key Findings

1. The composite primary outcome (severe bleeding, severe thromboembolism, or all-cause mortality at 6 months) occurred in 81% (87/107) of patients assigned to standard-dose UFH, 72% (78/108) assigned to low-dose UFH (absolute risk difference -9.1 percentage points [95% CI -20.3 to 2.1]), and 75% (79/105) assigned to LMWH.
2. Low-dose UFH and LMWH both met pre-specified statistical criteria for non-inferiority when compared to standard-dose UFH.
3. Severe bleeding occurred in 65% of the standard-dose UFH group, compared with non-significant numerical reductions to 58% in the low-dose UFH group and 59% in the LMWH group.
4. Severe thromboembolic complications did not increase with lower-intensity anticoagulation, occurring in 11% of the standard-dose UFH group, 10% of the low-dose UFH group, and 9% of the LMWH group.
5. Clear separation in anticoagulation intensity was achieved among the arms; the highest average daily aPTT was 68 seconds for standard-dose UFH, 56 seconds for low-dose UFH, and 38 seconds for LMWH.

Study Design

Design
Randomized Controlled Trial
Open-Label
Sample
330
Patients
Duration
6 mo
Median
Setting
7 Dutch ICUs
Population Adult patients (aged ≥18 years) requiring veno-venous (V-V) or veno-arterial (V-A) extracorporeal membrane oxygenation (ECMO) without a preexisting absolute indication for therapeutic anticoagulation.
Intervention Low-dose continuous unfractionated heparin (UFH) targeting an aPTT 1.5–2.0 times baseline OR therapeutic low-molecular-weight heparin (LMWH).
Comparator Standard-dose continuous unfractionated heparin (UFH) targeting an aPTT 2.0–2.5 times baseline.
Outcome A composite of severe bleeding during ECMO, severe thromboembolic complications during ECMO, or all-cause mortality at 6 months.

Study Limitations

The open-label design may have introduced performance and ascertainment biases, particularly regarding transfusion triggers or subjective adjudication of bleeding severity.
The standard-dose UFH target (aPTT 2.0–2.5 times baseline) may not reflect contemporary practices in all centers, as many institutions have empirically shifted to lower targets in recent years.
Anti-factor Xa monitoring practices were inconsistent across sites, potentially limiting the extrapolation of the findings to centers that heavily rely on anti-Xa rather than aPTT.
While it established non-inferiority, the trial may have been underpowered to detect small but clinically meaningful superiority in specific subgroups or secondary endpoints.

Clinical Significance

The RATE trial represents a pivotal shift in the management of extracorporeal membrane oxygenation (ECMO) by providing robust, randomized evidence that questions the long-standing dogma of requiring intense systemic anticoagulation. By proving that lower-intensity unfractionated heparin or low-molecular-weight heparin does not increase the risk of severe circuit or patient thrombosis—and concurrently trends toward reducing devastating bleeding complications—the trial strongly supports the adoption of permissive, lower-intensity anticoagulation strategies for adults undergoing ECMO.

Historical Context

During ECMO, blood is exposed to massive artificial circuit surface areas, inciting an intense pro-thrombotic cascade. Historically, critical care protocols extrapolated anticoagulation targets from cardiopulmonary bypass or deep vein thrombosis paradigms, utilizing high-intensity unfractionated heparin to prevent circuit failure or systemic embolization. However, major bleeding occurs in up to 30–60% of ECMO patients, heavily driving morbidity and mortality. Modern advancements in ECMO technology—such as heparin-bonded circuits, polymethylpentene oxygenators, and centrifugal pumps—have substantially improved hemocompatibility. Despite observational data hinting that lower anticoagulation targets were safer, rigorous guidelines remained absent due to a lack of powered randomized clinical trials. The RATE trial emerged as a landmark effort to officially compare these intensity strategies, redefining safety boundaries for ECMO anticoagulation.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

How do unfractionated heparin (UFH) and low-molecular-weight heparin (LMWH) differ in their mechanisms of action, and why might bleeding risk differ between them when used in an extracorporeal circuit?

Key Response

UFH inhibits both Thrombin (Factor IIa) and Factor Xa by binding antithrombin, whereas LMWH preferentially inhibits Factor Xa due to its shorter chain length. In an ECMO circuit, extensive contact activation occurs, making both pathways relevant. LMWH has more predictable pharmacokinetics, less protein binding, and a lesser effect on platelets, which theoretically reduces bleeding risk compared to standard-dose UFH.

Resident
Resident

When managing an ECMO patient randomized to the LMWH arm, what specific laboratory monitoring parameters are appropriate, and how do these differ from the monitoring required for standard-dose UFH?

Key Response

Standard UFH is typically titrated to a target aPTT (e.g., 50-70s) or anti-Xa (0.3-0.7 IU/mL). LMWH cannot be monitored with aPTT and requires LMWH-specific anti-Xa levels. Residents must recognize that monitoring LMWH in ECMO patients can be challenging due to altered pharmacokinetics and frequent concomitant acute kidney injury, which alters LMWH clearance and necessitates vigilant dose adjustments.

Fellow
Fellow

The RATE trial used a composite outcome of severe bleeding, severe thromboembolism, or 6-month mortality. How does the inclusion of 6-month mortality in this composite affect the interpretation of non-inferiority for anticoagulation strategies?

Key Response

Mortality in ECMO is highly driven by the underlying critical illness (e.g., ARDS, cardiogenic shock) rather than solely anticoagulation-related complications. Including a high-frequency, non-specific outcome like mortality in a composite endpoint can dilute the treatment effect regarding bleeding or thrombosis, mathematically biasing the trial toward non-inferiority. Fellows must critically assess whether the composite components are equally affected by the intervention.

Attending
Attending

Given the non-inferiority of low-dose UFH and LMWH in the RATE trial, how should we balance the logistical advantages of subcutaneous LMWH against its irreversibility and accumulation in acute kidney injury when selecting a strategy for the unit?

Key Response

LMWH reduces the need for continuous IV titrations and frequent lab draws, representing a major nursing and resource benefit. However, it is renally cleared (AKI is common in ECMO) and less completely reversible with protamine than UFH. Attendings must weigh these practical ward-management benefits against the necessity for rapid reversibility in cases of catastrophic cannula-site or intracranial bleeding.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

In an open-label, non-inferiority trial like RATE, what are the primary threats to validity regarding ascertainment bias and intent-to-treat (ITT) analyses, and how should a statistical analysis plan mitigate these?

Key Response

Open-label designs are susceptible to ascertainment bias, where investigators might underreport subjective bleeding or clotting events based on known allocations. Furthermore, in non-inferiority trials, ITT analyses tend to bias results toward the null (non-inferiority) due to non-adherence or crossover. A rigorous statistical plan must prioritize per-protocol analyses and utilize blinded, independent endpoint adjudication (PROBE design) to mitigate these threats.

Journal Editor
Journal Editor

What inherent methodological weaknesses in the RATE trial's open-label design and composite endpoint selection would a peer reviewer flag as potentially inflating the likelihood of declaring non-inferiority, and how does this impact editorial priority?

Key Response

A tough reviewer would flag that an open-label design combined with a composite endpoint heavily weighted by mortality creates a strong bias toward the null. If the event rate for mortality is high and unaffected by the drug, the margins for bleeding and clotting differences become blurred. This limits editorial enthusiasm unless the study proves that the non-inferiority margin was strictly preserved for the mechanism-specific endpoints (bleeding and thrombosis) independently.

Guideline Committee
Guideline Committee

Current ELSO (Extracorporeal Life Support Organization) guidelines generally recommend continuous UFH titrated to specific parameters. Based on the RATE trial findings, should ELSO guidelines be updated to recommend LMWH or low-dose UFH as a Class I alternative, or is this evidence insufficient?

Key Response

ELSO guidelines traditionally rely on continuous UFH due to its short half-life and reversibility, which is critical in dynamic ECMO patients. While the RATE trial provides RCT evidence that lower intensity or LMWH is non-inferior for a composite outcome, a guideline committee would likely assign it a Class IIa or IIb recommendation rather than Class I, pending confirmatory data in specific subsets (VA vs. VV ECMO) and addressing safety concerns regarding LMWH in patients with renal failure.

Clinical Landscape

Noteworthy Related Trials

2011

PROTECT Trial

n = 3,746 · NEJM

Tested

Dalteparin (LMWH) 5000 IU daily

Population

Critically ill medical-surgical ICU patients

Comparator

Unfractionated heparin (UFH) 5000 IU twice daily

Endpoint

Proximal deep-vein thrombosis

Key result: There was no significant difference in proximal DVT rates between the groups, but dalteparin significantly reduced the risk of heparin-induced thrombocytopenia.
2018

EOLIA Trial

n = 249 · NEJM

Tested

Early venovenous extracorporeal membrane oxygenation (ECMO)

Population

Patients with severe acute respiratory distress syndrome (ARDS)

Comparator

Conventional mechanical ventilation

Endpoint

60-day mortality

Key result: Early ECMO did not significantly lower 60-day mortality compared to standard care, but ECMO patients experienced a high incidence of severe bleeding events.
2021

ACTIV-4a / REMAP-CAP / ATTACC

n = 1,098 · NEJM

Tested

Therapeutic-dose anticoagulation (UFH or LMWH)

Population

Critically ill patients with severe COVID-19

Comparator

Standard-care prophylactic-dose anticoagulation

Endpoint

Organ support-free days up to day 21

Key result: Therapeutic-dose anticoagulation did not improve clinical outcomes and was associated with a significantly higher risk of major bleeding compared to standard prophylaxis.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis